128 



PROCEEDINGS OP THE AMERICAN ACADEMY. 



The only possible method of determining the amounts of these various 

 modifying eifects seems to be to vary the size of the exposed surface of 

 cathode; for upon tlie extent of this exposed surface depend the most 

 serious causes of error. This mode of experimentation was adopted by 

 Shaw, and especially by Gray in the paper just cited. Gray's results 

 are so interesting that it seems worth while to repeat a part of the plate 

 which depicts his observations. The abscissoe register the number of 

 square centimeters area per ampere, while the ordinates give the values 

 proportional to the amount of copper deposited by a given current in a 

 given time. The solution of cupric sulphate had a specific gravity vary- 

 ing from 1.15 to 1.18, or contained perhaps 250 grams of crystallized 

 salt to the litre. 



Area per ampere 



Figure 2. — Diagram drawn from the Numerical Table given by Gray. 



(Phil. Mag., [5], 25, 182 [1888].) 



It is evident that with large areas of cathode (small current densities) 

 the fairly straight lines tend to converge at a point somewhat above the 

 top of the table. On the other hand, when the current density becomes 

 great, the curves fail of their mark and bend sharply to the left. Gray 

 dismisses this remarkable tendency with the assumption of the "pro- 

 tective action " of strong currents already mentioned, and takes as the 

 true value the results obtained at 2° Centigrade with current densities 

 higher than 0.01 ampere per square centimeter. 



